CSR: Small: RUI: Octopus OS: A High-Performance Cloud OS with Accurate Resource Abstraction.

本项目旨在开发一种适应动态云环境的高性能云操作系统，称为Octopus OS。操作系统现在通常在现代云中的虚拟机(VM)中运行，而不是在物理机上运行，在现代云中，由于资源共享和VM迁移，资源高度动态。然而，传统的操作系统由于不了解资源的动态变化，不能有效地管理虚拟资源，导致云中资源利用率低，应用性能差。因此，拟议的八达通操作系统将创建和管理动态资源的准确抽象，并开发新的算法，以充分利用虚拟资源的独特特征，从而大大提高动态云中的应用程序性能和资源利用率。这项工作可以作为优化各种虚拟资源利用的框架来推动云操作系统的发展。结果将公布给公众使用。在项目开发期间将创建动手项目，以丰富现有的与Toos和云计算相关的课程。这个项目将包括来自代表性不足群体的学生和推动者为所有本科生研究计算机系统。将采取四个步骤来实现项目目标。首先，将在受控环境中分析主要虚拟化资源抽象(即CPU、内存和缓存)中的不匹配，以了解它们的影响，包括导致的系统异常行为和应用程序性能下降。其次，将创建准确的资源抽象，以使用探测技术修复已识别的不匹配。具体地说，将开发一组名为vProbers的微型基准测试，以分析虚拟化资源的动态性质，而不依赖于虚拟机管理程序支持或应用程序移植，使此解决方案在新兴的多云环境中切实可行。Octopus OS中探测到的准确资源抽象也可以被云应用程序和语言运行时利用，并且对于为跨云优化提供资源可见性至关重要。第三，新的抽象将向Octopus OS公开，以使其能够感知虚拟化。将开发新的算法来修复现有的系统异常行为，并释放动态资源的潜在性能优势。第四，将在各种公共云中进行资源要求高的基准测试，以评估建议的八达通操作系统适应多云环境中动态资源的能力。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This project aims to develop a high-performance cloud operating system (OS), called Octopus OS,that can adapt to the dynamic cloud environment. Instead of running on a physical machine,the OS now is often running in a virtual machine (VM) in the modern cloud where the resourcesare highly dynamic due to resource sharing and VM migration. However, the traditional OScannot manage the virtualized resources efficiently due to being unaware of the resource dynamics,leading to low resource utilization and poor application performance in the cloud. Therefore,the proposed Octopus OS would create and manage the accurate abstractions of the dynamicresources, and develop new algorithms to fully leverage the unique characteristics of the virtualizedresource, which would greatly improve application performance and resource utilization in thedynamic cloud. This work can advance the cloud OS by serving as a framework for optimizing theutilization of various virtualized resources. The outcomes will be released for public use. Hands onprojects will be created during the project development to enrich existing courses related toOS and cloud computing. This project will involve students from underrepresented groups and promoteresearch in computer systems for all undergraduates.Four steps will be taken to achieve the project goal. First, the mismatches in major virtualizedresource abstractions (i.e., CPU, memory, and cache) will be analyzed in a controlled environmentto understand their impacts, including resultant system abnormal behavior and applicationperformance degradation. Second, accurate resource abstractions will be created to fix the identifiedmismatches using probing techniques. Specifically, a set of micro-benchmarks, called vProbers,will be developed to profile the dynamic nature of the virtualized resources without relyingon the hypervisor support or application porting, making this solution practical in the emergingmulti-cloud environment. The accurate resource abstractions probed within Octopus OS can alsobe leveraged by cloud application and language runtime, and are critical to provide resource visibilityfor cross-cloud optimizations. Third, the new abstractions will be exposed to Octopus OS tomake it virtualization-aware. New algorithms will be developed to fix existing system abnormalbehavior and unlock the potential performance benefits of the dynamic resources. Fourth, experiments with resource-demanding benchmarks in various public clouds will be conducted to evaluate the ability of the proposed Octopus OS to adapt to the dynamic resources in the multi-cloud environment.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.